Accuracy versus migration overhead in real-time multiprocessor reweighting algorithms

We consider schemes for enacting task share changes - a process called reweighting - on real-time multiprocessor platforms. Our particular focus is reweighting schemes that are deployed in environments in which tasks may frequently request significant share changes. Prior work has shown that fair scheduling algorithms are capable of reweighting tasks with minimal allocation error. However, in such schemes preemption and migration overheads can be high. In this paper, we consider the question of whether the lower migration costs of partitioning-based schemes can provide improved average-case performance relative to fair-scheduled systems. Our conclusion is that partitioning-based schemes are capable of providing significantly lower overall error (including \´\´error" due to preemption and migration costs) than fair schemes in the average case. However, partitioning-based schemes are incapable of providing strong fairness and real-time guarantees